现代电子技术2024，Vol.47Issue(10)：159-163,5.DOI:10.16652/j.issn.1004-373x.2024.10.029
基于BP神经网络算法的超声电源频率追踪技术
Ultrasonic power frequency tracking technology based on BP neural network algorithm
摘要
Abstract
The ultrasonic vibration system is composed of the ultrasonic transducer and the power supply for ultrasonic composite processing.During the machining process,the ultrasonic transducer is influenced by various factors that can lead to resonance frequency drift.The ultrasonic composite processing power supply needs to output electrical signals corresponding to the frequency to ensure the stable operation of the ultrasonic vibration system.In order to avoid damage to the ultrasonic transducer,a control model based on BP neural network algorithm is designed,which can control the output frequency of the ultrasonic power supply by analyzing the operating history data of the ultrasonic power supply.The circuit simulation is conducted by means of Multisim,and the output signal of the ultrasonic composite processing power supply when the theoretical resonant frequency of the ultrasonic transducer is stable at 25 kHz is collected by the experiments.The Matlab simulation modeling is conducted to verify the control accuracy and reliability of the BP neural network model.The results show that the maximum error between the generated analog output and the actual output frequency does not exceed 5%,which is helpful for the stable operation of the ultrasonic composite processing power supply.关键词
BP神经网络/超声换能器/超声复合加工电源/频率追踪/电源控制/高频逆变Key words
BP neural network/ultrasonic transducer/ultrasonic composite processing power supply/frequency tracking/power control/high frequency inverter分类
信息技术与安全科学引用本文复制引用
惠红平,庄百亮,周永松,刘丁赫..基于BP神经网络算法的超声电源频率追踪技术[J].现代电子技术,2024,47(10):159-163,5.基金项目
江苏省产业前瞻与关键核心技术:竞争项目(BE2022106) （BE2022106）
江苏省产业前瞻与关键核心技术:碳达峰碳中和科技创新专项资金(BE2022032-3) （BE2022032-3）
